Reducing Bank Overdraft Usage through Price Discounts and SMS Reminders in Turkey

Fieldwork by:
108,000 bank clients
2011 - 2013
Target group:
  • Urban population
  • Adults
Outcome of interest:
  • Earnings and income
  • Credit balance/repayment
Intervention type:
  • Consumer protection
  • Information
  • Nudges and reminders
  • Monetary incentives
  • Pricing and fees
  • Simplification
AEA RCT registration number:

Because add-on charges for financial products, such as overdraft fees, can be complex and misleading, consumers may not fully understand the cost of using these additional services and may overuse them as a result. Researchers conducted a randomized evaluation to measure the impact of informational reminders and price discounts on bank clients’ overdraft usage in Turkey. While messages on the availability of overdraft protection services increased usage, messages promoting an overdraft discount actually reduced overdraft usage, potentially because the discount message reminded individuals of the costs associated with an overdraft. Overall, these results suggest that banks do not have incentives to reveal add-on prices of financial products, like overdraft credit, as it reduces demand.

Policy issue

Add-on charges for financial products, such as penalty fees, can be complex and misleading, thus misrepresenting the full price to the consumer. For example, many banks offer free checking accounts, but clients end up paying high fees when they overdraft, or make a transaction that leads their checking account to fall into negative balances. Overdraft credit offers clients additional funds to cover these transactions despite insufficient funds that must be repaid with interest, at a rate substantially higher than credit cards. Banks rarely provide information on the price of overdraft services - a major source of revenue for banks. Government regulators have increasingly scrutinized financial products pricing, but little rigorous evidence exists on why consumers use these products and incur the costs. Can SMS reminders and price discounts to inform consumers of overdraft fees reduce the use of these costly services?

Context of the evaluation

Turkey’s banking sector is well developed and innovative, with the highest percent of mobile banking users in Europe among internet users in 2013. However, many checking account holders are not approved for overdraft lines of credit because banks deem them too risky. Therefore, people who use overdraft services in Turkey are likely among the relatively rich, although likely not the richest, who have the wealth and access to cheaper credit that would make overdrafting unlikely. As in many countries, financial regulators in Turkey have been pursuing policies to make overdraft usage, pricing, and costs more transparent, for instance, by capping prices for overdraft services.

Researchers partnered with Yapi Kredi (YK), one of the five largest banks in Turkey’s retail banking sector. YK was interested in learning about the optimal strategy for pricing and advertising its overdraft service. Though study participants were selected due to infrequent overdraft activity, many did have some experience overdrafting; 18 percent of the sample overdrafted at least once between the four to twelve months before the evaluation. Participants were mostly male (71 percent), and over half were married (57 percent) and from the five largest cities in Turkey (51 percent).

Yapi Kredi

Details of the intervention

In partnership with Yapi Kredi (YK), researchers conducted several randomized evaluations to examine the impact of SMS messages offering reminders and price discounts on bank clients’ overdraft usage. The bank selected participants who were almost likely to overdraft but had not used the service in the prior three months. First, in August 2012, researchers randomly assigned half of the 108,000 existing checking account holders with overdraft services to receive a reminder about the availability of overdraft service, but not the cost. Then, in September, researchers sent each person in the sample one of six randomly selected messages

  • Overdraft availability reminder only: This message reminded customers that their account had overdraft services but provided no additional information on the cost.
  • Overdraft Interest Discount only: This message offered a 50 percent discount on all overdraft fees.
  • Automatic-Bill Pay Discount only: This message offered extra rewards points for authorizing automatic bill payments from their accounts. Customers could redeem reward points when making purchases with the account’s debit card.
  • Debt Card Discount only: This message offered extra rewards points for using their debit cards.
  • Overdraft Interest Discount with Automatic-Bill Pay Discount: This message offered the same discount on overdraft fees and extra rewards points for automatic bill payments.
  • Overdraft Interest Discount with Debt Card Discount: This message offered the same discount on overdraft fees and extra rewards points for using their debit cards.

Researchers also tested if the frequency or duration of the messages impacted client’s usage of YK’s products and services. While all participants received one message in September, afterwards, researchers randomly selected how often participants received additional SMS messages, either more frequently (every ten days), less frequently (every twenty days), or not at all during the evaluation. The messages also lasted longer for some randomly selected individuals, ending in November for some and in December for the others. Researchers used the bank’s administrative data to measure the impact of the various messages on customer’s overdraft and overall account usage six months after the end of the intervention.

Results and policy lessons

Messages promoting the overdraft discount reduced overdraft usage, especially when combined with other discounts on debit cards and automatic-bill payments, because it reminded consumers of the price of overdraft services. In contrast, consumers increased overdraft usage when they received messages only on the availability of overdraft services. Overall, these results suggest that that banks do not have incentives to reveal add-on prices of financial products, like overdraft credit, as it reduces demand.

Messages offering overdraft discounts reduced overdraft usage: Customers who received the messages with an overdraft discount were 1.2 percentage points less likely to have used overdraft services relative to those who received messages on the availability of overdraft. These messages also led to a reduction in the number of days that a client’s balance fell into overdraft. These results were driven by individuals who received messages on overdraft discounts coupled with discounts on automatic-bill pay or debit card use.

These results are somewhat surprising, given that generally discounts encourage greater demand. In fact, messages offering a discount for automatic-bill pay alone increased the likelihood a client registered for the service. Researchers suggest these counterintuitive effects may be due to consumers underestimating and being inattentive to the cost of overdraft services. With the discount messages, the bank revealed the cost of overdraft services and customers responded by reducing their demand for this expensive service.

In contrast, messages on the availability of overdraft services, without mention of price, increased usage: Compared to customers who received a message that did not mention any overdraft service, those who received a message on the availability of overdraft services increased the likelihood of using overdraft and the number of days that their balance fell into overdraft. This result illustrates that promoting overdrafts without mentioning the cost can increase demand. Customers may be unaware of all aspects of the overdraft service, not only its costs, and respond when the bank brings these aspects, like the availability, to their attention.

More frequent messages reinforced the messages’ impacts on consumer choices: For both the overdraft availability and discount messages, there was no impact on clients who received just one message, thus repeated messages drove the impact on consumer’s overdraft usage. Yet, there was no difference between receiving the messages every 10 days versus 20. Likewise, receiving the messages for an additional month had little influence on the effect of the messages. There was no evidence of long-term effects; none of these changes in behavior lasted beyond two months past the evaluation. The lack of long-term effects suggests that advertising or information campaigns must continue over time in order to be effective.

While reminders about these fees can lead to better consumer behavior, these changes in behavior do not last very long without sustained messaging, making information campaigns costly and presenting a difficult regulatory challenge. These results support policy maker’s increasing scrutiny of add-on pricing, but future research could help explain the welfare impact of similar policies to reveal the cost of add-on prices, in order to better inform policy decisions.

Alan, Sule, Mehmet Cemalcilar, Dean Karlan, and Jonathan Zinman. "Unshrouding Effects on Demand for a Costly Add-on: Evidence from Bank Overdrafts in Turkey." NBER Working Paper No. 20956, February 2015.